Video Coding Inter Prediction PDPC Simplification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently utilizing position-dependent prediction combination (PDPC) in inter prediction modes, particularly in terms of complexity and computational overhead, which affects coding efficiency and decoding speed.
Innovation Solution
The implementation of a simplified method for using PDPC in inter prediction, where the processing circuitry calculates intermediate values based on neighboring samples and combines them with inter prediction values, with specific weight calculations and boundary conditions to optimize prediction, and excludes certain neighboring blocks or modes based on block size and position requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If position-dependent prediction combination (PDPC) is applied in inter prediction mode, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the prediction process into distinct stages: generating intra prediction samples, calculating PDPC weights based on position, and combining predictions. This segmentation allows each stage to be optimized independently, reducing overall computational complexity while maintaining prediction accuracy.
Solution Approach 2:
The patent applies different prediction strategies to different regions within a block based on their position. PDPC is applied selectively to specific samples based on their location, using position-dependent weights to combine intra and inter predictions locally, rather than applying a uniform complex algorithm to the entire block.
2Measurement precision
If PDPC calculation is performed for all samples, then prediction precision is improved, but decoding speed decreases
Solution Approach 1:
The patent applies PDPC selectively to only certain samples within a block rather than all samples. The decision to apply PDPC to a sample depends on its position and boundary conditions, performing partial action only where it provides benefit, thus maintaining precision where needed while improving overall decoding speed.
Solution Approach 2:
The patent pre-calculates PDPC weights based on sample position before the actual prediction combination. These pre-computed weights are stored and reused, avoiding repeated complex calculations during decoding, which improves decoding speed while maintaining prediction precision.
3Measurement precision
If multiple neighboring blocks are used for PDPC, then prediction accuracy is improved, but processing overhead increases
Solution Approach 1:
The patent extracts and uses only the necessary neighboring samples required for PDPC calculation, rather than processing all available neighboring blocks. By selecting only the relevant neighboring samples based on position and availability, it reduces processing overhead while maintaining prediction accuracy.
Solution Approach 2:
The patent changes the parameters of neighboring block usage dynamically based on block position and boundary conditions. Different weights and selection criteria are applied to neighboring blocks depending on their position relative to the current block, optimizing the balance between accuracy and processing efficiency.
Data Source
AI summary
Aspects of the disclosure provide methods and apparatuses for video encoding/decoding. In some examples, an apparatus for video decoding includes receiving circuitry and processing circuitry. The processing circuitry decodes prediction information of a current block from a coded video bitstream. The prediction information is indicative of an inter prediction mode and a usage of a position dependent prediction combination (PDPC) in the inter prediction mode. Then, the processing circuitry calculates an intermediate value for a sample in the current block based on neighboring samples of the current block that are selected based on a position of the sample, and combines the intermediate value for the sample with an inter prediction value of the sample to reconstruct the sample.